The Eukaryotic UDP-N-Acetylglucosamine Pyrophosphorylases

A search of the yeast data base for a protein homologous to Escherichia coliUDP-N-acetylglucosamine pyrophosphorylase yieldedUAP1 (UDP-N-acetylglucosaminepyrophosphorylase), the Saccharomyces cerevisiae gene for UDP-N-acetylglucosamine pyrophosphorylase. The Candida albicans and human homologs were also cloned by screening a C. albicans genomic library and a human testis cDNA library, respectively. Sequence analysis revealed that the human UAP1 cDNA was identical to previously reported AGX1. A null mutation of the S. cerevisiae UAP1 (ScUAP1) gene was lethal, and when expressed under the control of ScUAP1 promoter, bothC. albicans and Homo sapiens UAP1(CaUAP1 and HsUAP1) rescued theScUAP1-deficient S. cerevisiae cells. All the recombinant ScUap1p, CaUap1p, and HsUap1p possessed UDP-N-acetylglucosamine pyrophosphorylase activitiesin vitro. The yeast Uap1p utilizedN-acetylglucosamine-1-phosphate as the substrate, and together with Agm1p, it produced UDP-N-acetylglucosamine from N-acetylglucosamine-6-phosphate. These results demonstrate that the UAP1 genes indeed specify eukaryotic UDP-GlcNAc pyrophosphorylase and that phosphomutase reaction precedes uridyltransfer. Sequence comparison with other UDP-sugar pyrophosphorylases revealed that amino acid residues, Gly112, Gly114, Thr115, Arg116, Pro122, and Lys123 of ScUap1p are highly conserved in UDP-sugar pyrophosphorylases reported to date. Among these amino acids, alanine substitution for Gly112, Arg116, or Lys123 severely diminished the activity, suggesting that Gly112, Arg116, or Lys123 are possible catalytic residues of the enzyme.

UDP-N-acetylglucosamine (UDP-GlcNAc 1 ) is a ubiquitous and essential metabolite and plays important roles in several metabolic processes. In bacteria, it is known as a major cytoplasmic precursor of cell wall peptide glycan and the disaccharide moiety of lipid A (1)(2)(3). In eukaryotes, it serves as the substrate of chitin synthase, whose product is shown to be essential for fungal cell wall (4). It is also used in the GlcNAc moiety of N-linked glycosylation and the GPI-anchor of cellular proteins (5).
On the other hand, there are three UDP-sugar pyrophosphorylase genes in S. cerevisiae reported to date. GAL7 (21) and UGP1 (22) encode UDP-galactose (UDP-Gal) pyrophosphorylase and UDP-glucose (UDP-Glc) pyrophosphorylase, respectively. Recently, VIG9 was identified as the GDP-mannose (GDP-Man) pyrophosphorylase gene by functional complementation using the glycosylation defective vig9-1 mutant (23), and the possible amino acid sequence motif for the active site of UDP-sugar pyrophosphorylase is proposed. Because all of these enzymes preserve substrate specificity to a certain type of sugar, there should be an enzyme specific to GlcNAc-1-P.
In an attempt to identify the gene for UDP-GlcNAc pyrophosphorylase, we searched the S. cerevisiae genome data base and found that the protein specified by YDL103C. The Candida albicans and human homologs were also isolated and characterized. From sequence comparison and mutation analysis, the probable catalytic residues of UDP-sugar pyrophosphorylases are proposed. * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
The nucleotide sequence(s) reported in this paper has been submitted to the GenBank TM /EBI Data Bank with accession number(s) AB011272, AB011003, AB011004.

EXPERIMENTAL PROCEDURES
Yeast Data Base Search and Screening of DNA Libraries-An amino acid sequence motif of LXXGXGTXMXXXXPK where X represents any amino acid was obtained by comparing the amino acid sequences of E. coli GlmUp (EcGlm1p), S. cerevisiae Ugp1p (ScUgp1p), and Homo sapiens Ugp1p (HsUgp1p), and was used to search the S. cerevisiae genome data bases. The entire open reading frame of ScUAP1 (originally designated YDL103C) was amplified by polymerase chain reaction using the S. cerevisiae genomic DNA extracted from strain A451 (MAT␣ can1, aro7, can1, leu2, trp1, ura3) as a template, and cloned at the XbaI site of pUC18 or pYEUra3 (Toyobo) generating pUC-ScUAP1 and pYEU-ScUAP1, respectively. Primers used for polymerase chain reaction were 5Ј-AGATCTAGAATGACTGACACAAAACAGCT-3Ј and 5Ј-AGATCTAGATTATTTTTCTAATACTATAC-3Ј.
The C. albicans and human homologs of ScUAP1 were cloned by screening a C. albicans genomic DNA library and a human testis cDNA library using the 1.4-kilobase EcoRI-EcoRI fragment of ScUAP1 as a probe. Hybridization and washing of the filters were carried out under stringent conditions (20 mM sodium phosphate (pH 7.2), 5ϫ SSC (1ϫ SSC contains 150 mM NaCl and 15 mM sodium citrate), 5ϫ Denhardt's solution, 0.1% SDS, 25% formamide at 42°C for hybridization; 0.1ϫ SSC and 0.1% SDS at 50°C for washing). Bacterial cells and phages that were strongly hybridized with the probe DNA were collected. After the third screening, DNA was extracted from bacterial cells and phages, and the insert DNA was cloned at the SmaI site of pUC19 for further plasmid construction. Radiolabeling of the probe DNA was performed by the random priming method using [␣-32 P]dCTP (24), and DNA sequencing was carried out as described elsewhere (24). Construction of the C. albicans genomic DNA library was already reported (25). A human testis cDNA library was purchased from CLON-TECH (USA).
Expression and Purification of the Recombinant Proteins-The coding regions of ScUAP1, CaUAP1, HsUAP1, and ScAGM1 were cloned at the EcoRI (for ScUAP1 and ScAGM1) or SmaI (for CaUAP1 and HsUAP1) site of pGEX2T (26), and the resulting plasmids were transfected into E. coli JM109 to let them express recombinant yeast and human proteins as a fusion product with glutathione S-transferase (GST). Induction and expression of the recombinant Uap1 proteins was carried out with isopropyl ␤-D-thio-galactopyranoside as described (25,26). At 4 h after the addition of isopropyl-␤-D-thio-galactopyranoside, the bacterial cells were harvested, suspended in a buffer containing 20 mM Tris-HCl (pH 7.5), 0.5 mM EDTA, 50 mM NaCl, 10 mM ␤-mercaptoethanol, 10%(v/v) glycerol, 1 mM phenylmethylsulfonyl fluoride, and lysed by sonication. After cell debris were removed by centrifugation at 15,000 ϫ g at 4°C for 30 min, GST-Uap1 and GST-Agm1 fusion proteins were purified by glutathione Sepharose CL-4B column chromatography, as described (26) and analyzed by sodium dodecyl sulfatepolyacrylamide gel electrophoresis. The primers used for amplifying the ScAGM1 open reading frame (ORF) were 5Ј-CGGGAATTCATAAGGT-TGATTACGAGCAAT-3Ј and 5Ј-ATTGAATTCTCAAGCAGATGCCTT-AACGTG-3Ј.
Assays for UDP-GlcNAc Pyrophosphorylase-An assay for UDP-GlcNAc pyrophosphorylase was performed in a 20 l standard reaction mixture containing 50 mM Tris-HCl (pH 8.3), 5 mM MgCl 2 , 20 M GlcNAc-1-P, 10% (v/v) glycerol, and 0.1 M [␣-32 P]UTP (specific activity 1 ϫ 10 3 cpm/pmol) and approximately 0.1 g of the indicated recombinant proteins at 30°C for 10 min. 2 l of each reaction mixture were spotted onto polyethyleneimine cellulose plates, and nucleotide sugars were separated by thin layer chromatography (TLC) in a solution that was prepared by mixing 6 g of Na 2 B 4 O 7 /10H 2 O, 3 g of H 3 BO 3 , and 25 ml of ethylene glycol in 70 ml of H 2 O (27). The radioactive spots were visualized by autoradiography. An alternative high flux assay was carried out in 90 l of reaction mixture containing 50 mM Tris-HCl (pH 8.3), 5 mM MgCl 2 , 25 M UTP, 20 M GlcNAc-1-P, 10% (v/v) glycerol, 1 mM dithiothreitol, 0.4 units/ml pyrophosphatase (Sigma), and approximately 0.1 g of the recombinant enzyme. After incubation at 30°C for 10 min, 100 l of the color reagent containing 0.03% (w/v) malachite green, 0.2% (w/v) ammonium molybdate, and 0.05% (v/v) Triton X-100 in 0.7 N HCl was added to the reaction mixture, which was followed by incubation at room temperature for 5 min. Inorganic phosphate derived from the pyrophosphate and thereby representing the enzyme activity was quantified by measuring optimal density at 655 nm.
The S. cerevisiae agm1⌬ null mutant strain was obtained by a means similar to that for the ScUAP1 depletion. The entire ORF of ScAGM1 was cloned at the XbaI site of pUC18 and pYEUra3, generating pUC-ScAGM1 and pYEU-ScAGM1, respectively. The 1.4-kilobase BalI-BglII region of the ScAGM1 ORF in pUC-ScAGM1 was replaced by LEU2, generating pUC-AGM1L. YPH499 cells were transformed with pYEU-ScAGM1 and then with pUC-ScAGM1L that had been previously digested with XbaI. The resulting ura ϩ leu ϩ transformants, which grew in galactose medium but died in glucose medium, were collected and used as agm1⌬ strain (MATa ura3, lys2, ade2, trp1, his3, leu2, agm1⌬::LEU2 AGM1-URA3).
To test the ability of CaUAP1, HsUAP1, and the mutant ScUAP1 to complement ScUAP1, the entire ORFs of CaUAP1, HsUAP1, and the mutant ScUAP1 were cloned in pRS414 -1 where a 2.0-kilobase BglII-XbaI fragment encompassing the ScUAP1 promoter was inserted at the BamHI site of pRS414 (Stratagene). Thus, the transcription of CaUAP1 and HsUAP1 from this plasmid was under the control of the ScUAP1 promoter. The resulting plasmids were transfected into uap1⌬ cells. After selection of trp ϩ cells in the presence of galactose, they were transferred to plates containing glucose and further cultured for 3 days.
Site-directed Mutagenesis-A series of the ScUAP1 mutants harboring an alanine substitution for Gly 111 , Gly 112 , Gly 114 , Thr 115 , Arg 116 , Leu 117 , Pro 122 , or Lys 123 were generated by the oligonucleotide-directed dual amber method as described (28) with Mutan-Express Km TM (Takara). The entire ORF of the ScUAP1 gene was cloned at the EcoRI site of pKF18k (Takara) using EcoRI linker and hybridized with oligonucleotides containing the indicated mutations. The resulting mutant ScUAP1 genes were excised from the vector and ligated at the EcoRI site of pGEX-2T and the BamHI site of pRS414 -1. All the mutations were confirmed by sequencing the DNA.

Cloning of the Yeast UDP-GlcNAc Pyrophosphorylase Gene-
Three distinct UDP-sugar pyrophosphorylase activities are present in yeast. In S. cerevisiae, the GAL7 (21), UGP1 (22), and VIG9 (23) genes have been shown to encode UDP-Gal pyrophosphorylase, UDP-Glc pyrophosphorylase, and GDP-Man pyrophosphorylase, respectively, but the gene for UDP-GlcNAc remains to be established. Comparison of the amino acid sequences between E. coli UDP-GlcNAc pyrophosphorylase (GlmUp) and S. cerevisiae UDP-Glc pyrophosphorylase (Ugp1p) identified an amino acid sequence motif, L(X) 2 GXGT-XM(X) 4 PK, where X represents any amino acid. In an attempt to identify the S. cerevisiae UDP-GlcNAc pyrophosphorylase gene, we searched the yeast data base and found that PSA1 and YDL103C could encode proteins with a sequence similar to the above amino acid motif (Fig. 1). PSA1 is identical to VIG9, which has been shown to be the GDP-Man pyrophosphorylase gene. Accordingly, we asked whether YDL103C specifies UDP-GlcNAc pyrophosphorylase. The Ydl103c protein was expressed in E. coli as a fusion protein with GST and purified by affinity column chromatography using glutathione-Sepharose CL-4B. The purified GST-Ydl103c fusion protein produced [ 32 P]UDP-GlcNAc when incubated with GlcNAc-1-P and [␣-32 P]UTP, whereas GST alone did not (Fig. 2). The above result demonstrates that YDL103C is a gene for UDP-GlcNAc pyrophosphorylase, and, therefore, the gene was designated ScUAP1 (the S. cerevisiae UDP-GlcNAc pyrophosphorylase gene 1).
Because uridyltransfer to GlcNAc-1-P releases pyrophosphate from UTP, we developed a conventional high-flux assay by adding pyrophosphatase to the reaction mixture, which allows us to estimate the enzyme activity from the amounts of inorganic phosphates produced after the hydrolysis of pyro-phosphates in the reaction mixture. By this assay, it was demonstrated that the GST-ScUap1 fusion protein converted Glc-NAc-1-P to UDP-GlcNAc in a dose-dependent manner (see below). Moreover, UTP was essential for the production of UDP-GlcNAc by ScUap1p; none of ATP, GTP, and CTP were used as the substrate (not shown).
Because UDP-GlcNAc is an essential metabolite serving as a precursor of cell wall chitin, protein N-glycosylation, and GPI anchor in yeast (4,5), ScUAP1 may be an essential gene for viability if it is the only UDP-GlcNAc pyrophosphorylase gene in S. cerevisiae. The S. cerevisiae uap1⌬ null mutant strain in which the endogenous UAP1 gene was disrupted, but where episomal copies of UAP1 whose transcription was under the control of GAL1 promoter were maintained, grew on galactose plates but died on glucose plates. The cells of S. cerevisiae uap1⌬ null mutant displayed an aberrant morphology; most of the yeast cells fully swelled and some were lysed, which is a phenotype quite similar to that caused by a null mutation of AGM1, the gene for GlcNAc phosphate mutase (Fig. 3). This is suggestive that the ScUAP1 is a sole UDP-GlcNAc pyrophosphorylase gene in S. cerevisiae and that the most apparent defect resulting from depletion of the functional UAP1 occurred in the cell wall.
Identification of the UDP-GlcNAc Pyrophosphorylase Genes of Other Organisms-To gain more insight into the characteristics of UDP-GlcNAc pyrophosphorylase, we intended to isolate the ScUAP1 homologs from the pathogenic fungus C. albicans as well as from human. By screening a C. albicans genomic DNA library and a human testis cDNA library with ScUAP1 DNA as a probe, CaUAP1, and HsUAP1, C. albicans (Ca) and the human (Hs) homologs of ScUAP1, were cloned and sequenced. The predicted products of ScUAP1, CaUAP1, and It should be noted that UDP-Glc, which appeared in the presence of ScUap1p and Glc-1,6-P 2 , was formed from a trace of Glc-1-P in the Glc-1,6-P 2 . B, the indicated amounts of the purified GST-ScAgm1p were incubated with approximately 0.1 g of the purified GST-ScUap1p, GlcNAc-6-P, and UTP in the presence (E) or absence (‚) of 20 mM Glc-1,6-P 2 . The amounts of the released inorganic phosphate that represent the enzyme activities were determined with malachite green and ammonium molybdate.
HsUAP1 are highly related to each other (Fig. 1). Interestingly, the cloned HsUAP1 cDNA was identical to the previously reported AGX1 cDNA whose product is implicated as being an antigen causing male infertility (29). Both of the recombinant CaUap1p and HsUap1p, which were expressed in E. coli as a fusion with GST, possessed UDP-GlcNAc pyrophosphorylase activities (Fig. 2), confirming that CaUAP1 and HsUAP1 indeed specify UDP-GlcNAc pyrophosphorylase. Furthermore, expression of CaUAP1 or HsUAP1 under the control of the ScUAP1 promoter supported the growth of the UAP1-deficient S. cerevisiae cells even in the presence of glucose. Thus, it appears that both the C. albicans and human UAP1 functionally complement ScUAP1 (Fig. 4).
Substrate Specificity of UDP-GlcNAc Pyrophosphorylase-We next examined the substrate specificity of UDP-Glc-NAc pyrophosphorylase using ScUap1p. ScUap1p reproducibly converted GlcNAc-1-P into UDP-GlcNAc in the presence of UTP but did not utilize GlcNAc-6-P, galactose-1-phosphate (Gal-1-P) or mannose-1-phosphate (Man-1-P) as a substrate (Fig. 5A). Unexpectedly, the enzyme generated a spot whose mobility corresponded to that of UDP-Glc from Glc-1-P, indicating the dual substrate utility of Uap1p. However, Glc-1-P was much less efficient as shown in Fig. 5B. Consequently, ScUAP1 did not complement ScUGP1 (data not shown).
It is believed that the interconversion of GlcNAc-6-P and GlcNAc-1-P precedes the uridyltransfer in vivo. This prompted us to add the yeast GlcNAc phosphate mutase (ScAgm1p) to the reaction mixture. As shown in Fig. 6A, together with ScAgm1p, ScUap1p produced UDP-GlcNAc from GlcNAc-6-P, whereas ScAgm1p alone did not. It is also demonstrated that hexosephosphate mutases require Glc-1,6-P 2 either as an activator or a cofactor for the catalytic reaction (30,31). Therefore, we examined the effect of Glc-1,6-P 2 on the synthesis of UDP-GlcNAc from GlcNAc-6-P. The TLC analysis of the products indicated that Glc-1,6-P 2 was not essential for the interconversion of GlcNAc-6-P and GlcNAc-1-P, because UDP-GlcNAc was produced from GlcNAc-6-P by Agm1p and Uap1p even in the absence of Glc-1,6-P 2 (Fig. 6A). However, further assessment of the importance of Glc-1,6-P 2 for the reaction by ScAgm1p revealed that the enhancement by Glc-1,6-P 2 of the mutase reaction was significant when the ScAgm1p concentration was rather low (Fig. 6B).
Possible Active Sites of ScUap1p-Comparison of the amino acid sequences among UDP-sugar pyrophosphorylases revealed that the region between amino acid positions 111 and 123 of ScUap1p shares significant sequence identity with other UDP-sugar pyrophosphorylases (Fig. 7). To verify the impor-tance of this region for the catalytic activity, the highly conserved amino acids in this region, Gly 111 , Gly 112 , Gly 114 , Thr 115 , Arg 116 , Leu 117 , Pro 122 , and Lys 123 were replaced by alanine. As was done for the wild type ScUap1p, all the mutant enzymes were expressed as a fusion with GST and purified by affinity column chromatography (Fig. 8A). Although Gly 114 , Thr 115 , and Pro 122 are also highly conserved in known UDP-sugar pyrophosphorylases, replacement of these amino acids by alanine only weakly impaired the enzyme activity. In contrast, substitution of alanine for Gly 112 , Arg 116 , or Lys 123 severely diminished the activity (Fig. 8B). Furthermore, G112A but not other mutants displayed a higher K m value to GlcNAc-1-P (Table I), and all of G112A, R116A, and K123A failed to rescue the S. cerevisiae uap1⌬ null mutant (data not shown). None of the mutations significantly affected the K m values in response to UTP (Table I). Taken together, it was proposed that Gly 112 serves as a binding site for GlcNAc-1-P, and that Gly 112 , Arg 116 , and Lys 123 are possible catalytic residues.

TABLE I Characteristics of the mutant ScUap1p
The K m and k cat values of the wild type and mutant enzymes to GlcNAc-1-P were determined from the amounts of pyrophosphate released from UTP. The K m values to UTP were also indicated in the right column.  FIG. 8. Effects on ScUap1p activity of alanine substitution for the conserved amino acids. The ScUap1 mutant proteins harboring an alanine substitution for each of the amino acids that are highly conserved in UDP-sugar pyrophosphorylases were expressed as a fusion with GST and purified with glutathione-Sepharose beads. A, approximately 1 g of the wild type and the indicated mutant proteins were separated on a 10% SDS-polyacrylamide gel (PAGE) and stained with Coomassie Brilliant Blue. The position of GST-ScUap1p is indicated by the arrowhead. B, approximately 0.1 g of the purified GST and the indicated mutant proteins were incubated with GlcNAc-1-P and UTP, and the amounts of the released inorganic phosphate that represent the enzyme activities were determined with malachite green and ammonium molybdate.

DISCUSSION
In this paper, we have identified the eukaryotic UDP-GlcNAc pyrophosphorylase genes. The expected amino acid sequences of the yeast and human enzymes are well conserved, and both C. albicans and human enzymes functionally complement S. cerevisiae UAP1. Although the yeast enzyme catalyzed uridyltransfer to Glc-1-P, ScUap1p displayed a reasonable substrate specificity to GlcNAc-1-P, because the enzyme utilized Glc-1-P much less efficiently than GlcNAc-1-P. In fact, overexpression of ScUAP1 did not overcome the lethal phenotype caused by a depletion of UGP1 in S. cerevisiae. Moreover, the enzyme did not recognize GlcNAc-6-P, but together with ScAgm1p, it produced UDP-GlcNAc from GlcNAc-6-P, demonstrating that the GlcNAc phosphate mutase reaction precedes uridyltransfer in UDP-GlcNAc biosynthesis. However, we cannot rule out the possibility that the results of the TLC assays and the high flux assays may not be exactly the same, because in the high flux assay the reverse reaction was eliminated by pyrophosphatase.
Both UDP-GlcNAc pyrophosphorylase and GlcN-1-P acetyltransferase activities are authentic in E. coli GlmUp (9). It has also been demonstrated that the N-terminal region is responsible for the uridyltransfer, and acetylase activity resides in the C-terminal half of GlmUp (13). Unlike bacterial UDP-GlcNAc pyrophosphorylase, the eukaryotic enzymes seem not to be bifunctional, because ScUap1p did not utilize GlcN-1-P as the substrate and the C-terminal portion of GlmUp showed no significant sequence homology to any UDP-GlcNAc pyrophosphorylase. Thus, it is likely that in yeast, GlcN-6-P is first acetylated by an as yet unidentified enzyme and then the mutase reaction generates GlcNAc-1-P.
Phosphomannomutase and phosphoglucomutase require a sugar biphosphate as a cofactor, which serves as a phosphate donor necessary to activate the enzyme by phosphorylation (30,31). In this study, ScAgm1p was able to produce GlcNAc-1-P even in the absence of cofactor, Glc1,6-P 2 , if a sufficient amount of ScAgm1p was present. One possible explanation for this discrepancy is that a small portion of the recombinant ScAgm1p was already phosphorylated and thereby activated. However, this hypothesis is inconsistent with the recent report by Oesterhelt et al. (31) that the plant and yeast enzymes utilize a sugar diphosphate as a co-substrate.
Sequence comparisons of the UDP-sugar transferases revealed that there is a region where the amino acid sequence is highly conserved among most of the known UDP-sugar pyrophosphorylases. Alanine substitution for Gly 112 , Arg 116 , or Lys 123 severely diminished the enzyme activity and ability to complement the wild type ScUAP1 gene, strongly suggesting that these amino acids are catalytic residues. Among these three amino acids, Gly 112 was shown to be a possible binding site to GlcNAc-1-P, because G112A displayed an increased K m value. In human UDP-Glc pyrophosphorylase, it was demonstrated that a single mutation of Gly 115 to Asp drastically impaired the enzyme activity and caused cellular UDP-Glc deficiency (32). Sequence comparison of the Uap and Ugp proteins reveals that Gly 115 of the human Ugp1p corresponds to Gly 112 of the yeast Ugp1p. Thus, it is likely that Gly 115 of the human Ugp1p also serves as a Glc-1-P binding site. Curiously, Gal7p, which is known as UDP-Gal pyrophosphorylase, shares no significant sequence homology to known UDP-sugar pyrophosphorylases, and the conserved amino acids essential for the catalytic activity of ScUap1p are not found in Gal7p (21). This may imply that the catalytic mechanism of Gal7p differs from those of other UDP-sugar pyrophosphorylases.
The human UDP-GlcNAc pyrophosphorylase cDNA turned to be identical to the AGX1 cDNA. Although the physiological function of AGX1 remains to be established, it encodes an unknown antigen expressed in infertile males and is implicated in antibody-mediated human infertility (29). AGX1 is abundantly expressed in testes, and only low levels of AGX1 mRNA were detected in placenta, muscle, and liver (29). The reason why testis expresses a higher level of AGX1 mRNA and how UDP-GlcNAc pyrophosphorylase causes human male infertility await further study. In addition, there is an additional AGX cDNA, termed AGX2, which differs from AGX1 by a 48-base pair insertion in the ORF. The level of AGX2 mRNA was not remarkably increased in testis; low but similar levels of AGX2 mRNA were detected in testis, placenta, muscle, and liver (29). Therefore, it may also be of interest to study how the internal 48-base pair insertion affects the UDP-GlcNAc pyrophosphorylase activity.